4 ms·

nothing about data centres in space makes sense if you understand even a modicum of physics https://taranis.ie/datacenters-in-space-are-a-terrible-horrible-no-g

by yogthos 5mo ago

nothing about data centres in space makes sense if you understand even a modicum of physics https://taranis.ie/datacenters-in-space-are-a-terrible-horrible-no-good-idea/ https://taranis.ie/datacenters-in-space-are-a-terrible-horri...

OutOfHere 5mo ago
The primary argument of the article is that cooling in space is limited, that GPUs emit significant heat, that this is the primary blocker. Well, as it stands, datacenters aren't effectively cooling on the Earth. They're passing their heat to the planet, and are heating up the planet. They're producing significant CO2 emissions, and using up the water and electricity resources of communities. The problems from having datacenters on Earth far exceed having them in space. In space, it's a closed small system, so the equations are immediately apparent. On Earth we just make it the next generation's problem. The ISS is constrained by being in Earth orbit. There however is absolutely no reason for datacenters to be in Earth orbit, and so they have much larger space available to them for radiative cooling while being in solar orbit. As for the hardware, it cannot be what we use on Earth. It has to be hardened chips for use in space. If you lack an imagination, it's your problem, and societally it will cost us everything. I advise scrolling down on your linked page to read the comments by Saha, etc.
- yogthos 5mo ago
  The primary argument of the article is that it's physically impossible to dissipate enough heat to power even a modest data centre in space. What you refer to as lack of imagination is my understanding of basic thermodynamics which you sadly lack.
  - OutOfHere 5mo ago
    There is no size limit to radiators in space. Also, the radiative surface can very well be in 3D, just not packed as densely as a heatsink for use on Earth.
  - OutOfHere 5mo ago
    Secondly, there is a lot that can be done to drive down the size of the radiator. A heat pump can be used to radiate heat at much higher temperatures, thereby requiring a smaller radiator. Moreover, a metamaterial surface can be used to prevent self-absorption. On Earth, people like you keep sending the bill to the next generation. In space, the bill is due immediately. There is no real difference in the amount due.
  - OutOfHere 5mo ago
    There is a video (AI researched and generated) that discusses heat rejection for space compute, mainly achievable by raising the temperature of heat rejection: https://www.youtube.com/watch?v=s7Mv_OcBXI8 https://www.youtube.com/watch?v=s7Mv_OcBXI8

by yogthos 5mo ago

OutOfHere 5mo ago

The primary argument of the article is that cooling in space is limited, that GPUs emit significant heat, that this is the primary blocker. Well, as it stands, datacenters aren't effectively cooling on the Earth. They're passing their heat to the planet, and are heating up the planet. They're producing significant CO2 emissions, and using up the water and electricity resources of communities. The problems from having datacenters on Earth far exceed having them in space. In space, it's a closed small system, so the equations are immediately apparent. On Earth we just make it the next generation's problem. The ISS is constrained by being in Earth orbit. There however is absolutely no reason for datacenters to be in Earth orbit, and so they have much larger space available to them for radiative cooling while being in solar orbit. As for the hardware, it cannot be what we use on Earth. It has to be hardened chips for use in space. If you lack an imagination, it's your problem, and societally it will cost us everything. I advise scrolling down on your linked page to read the comments by Saha, etc.

yogthos 5mo ago
The primary argument of the article is that it's physically impossible to dissipate enough heat to power even a modest data centre in space. What you refer to as lack of imagination is my understanding of basic thermodynamics which you sadly lack.
- OutOfHere 5mo ago
  There is no size limit to radiators in space. Also, the radiative surface can very well be in 3D, just not packed as densely as a heatsink for use on Earth.
- OutOfHere 5mo ago
  Secondly, there is a lot that can be done to drive down the size of the radiator. A heat pump can be used to radiate heat at much higher temperatures, thereby requiring a smaller radiator. Moreover, a metamaterial surface can be used to prevent self-absorption. On Earth, people like you keep sending the bill to the next generation. In space, the bill is due immediately. There is no real difference in the amount due.
- OutOfHere 5mo ago
  There is a video (AI researched and generated) that discusses heat rejection for space compute, mainly achievable by raising the temperature of heat rejection: https://www.youtube.com/watch?v=s7Mv_OcBXI8 https://www.youtube.com/watch?v=s7Mv_OcBXI8